TythosEternal

To crush your Kerbals' hopes! To see them driven before you! To hear the lamentations of their boosters!

There is a mental NASA risk spectrum map floating around in my head, in which (from least to most risk) the order is: Boeing, SpaceX, SNC. I think it's most likely, given their process and priorities, that the selection will be made based off of a risk-versus-cost frontier in which NASA seeks to minimize both at the expense of novelty, capability, and diversity. In a perfect world, I think NASA would choose two and bookend the spectrum by selecting Boeing and SNC. I am a big fan of "portfolio diversification," if you will, and if I were in charge, I'd want to maximize the experience we have with the advantage and disadvantages of both approaches. There's clear tradeoffs, but if a lifting body design can be proven to sufficiently reduce the weight, re-entry, and complexity issues that plagued the space shuttle, I think it will be a major step forward. The pessimist in me thinks that, despite the great success of CRS, NASA will choose only one supplier for CTS. In this case, it is virtually guaranteed to be a Boeing-versus-SpaceX faceoff, and the evaluation of launch vehicle risk (well, and the political question, which I will only mention briefly here) will likely be the determining factor. While there is the spectre of RD-180 dependence and the risk associated with the evolution of an alternative, it is naive to think that NASA will blindly accept a Falcon 9 / Dragon v2 stack with open arms. It's not easy to quantify, but there is still a great deal of skepticism, cynicism, and resentment within industry (both NASA and private) towards SpaceX. Common complaints range from "they're just copying our work and low-balling the cost," to "they're an upstart bunch of undergrads who have gotten lucky but have no real appreciation of risk," to the usual resentment you see against novel and successful risk-takers. Certain organizations will go far, far out of their way to ensure Musk never sees a dime of their money. If a substantial part of the selection board reflects this attitude, you can be sure to expect some rationalization of the Atlas V risk to accompany the decision. Add to this the political clout and relationships Boeing and ULA maintain with all levels of NASA, its congressional watchdogs, and other industry associates, and it looks like it will really take some unusually honest balls for NASA to sole-source this to SpaceX.

To me, there are two strong general approaches that can be taken that ensure relevancy to the course material. The first is an applied physics approach, in which you look at ways basic physical laws create complex and interesting systems (even when simplified to two bodies). Apply as much math to this approach as desire and/or appropriate. The second approach is a more engineering-oriented concept, in which you treat KSP as an opportunity to get students thinking about the design and experimentation process. Here, you sacrifice emphasis on the physical laws and instead give them specific challenges to solve from a reasonable starting point. I can think of other approaches--and many have been named on this thread--but having been both a student and instructor of AP Physics class, I strongly urge you to keep time considerations on the top of your head. Focus on the approaches that will give the best fun-lesson-for-time ratio, rather than sharing as much of KSP as possible (while tempting!). I know we're all really looking forward to hearing how it goes! Please keep us updates regarding your efforts.

I'm a big fan of the physical reading experience. My wife puts up with my book collecting (my 1898 Kipling is my baby), though I've always thought it's much more durable and practical than her equally-feverish lust for shoes. One of my big fantasies, when I strike it rich, is to build a big house in the coastal mountains of Oregon that would be nothing but a huge library with some bedrooms, bathrooms, and a kitchen off to the side. Similarly, I'm always reminding my daughters that Belle is the best Disney princess, because she loves to read and gets that kick-ass library. Why do I bring these random things up? It is impossible to have the same experiences with an e-book, no matter how good the information density. That having been said, I'd love to get a Nook one of these days for traveling, low-light conditions, and other situations where physical books just aren't practical. It will always be a fallback, though, and never a replacement. You know what would be awesome? Barnes and Noble (where I drop about half my expendable income on a regular basis) should, for members, give you a free electronic copy of each physical book you purchase, sort of like how movie publishers are finally letting you access digital copies of films when you but the physical disc.

I first picked up EVE back in 2004, and still return to it periodically. It's a fun game, and I really appreciate how scalable the appeal is: if you only have a few hours a week, you can still enjoy mining and manufacturing with minimal impact of lost time, thanks to the skill training model. It's also more flexible, in that (outside of null sec, at least) you can start mining, go change your kid's diaper, flip the burgers for dinner, and return in time to refine. Of course, there are much more hardcore ways to play. My corp has never really scale up that much, but all of the devious territorial struggles have always fascinated me. It would be fun, I'm sure, I just don't have that kind of time.

Just so there's a direct link... http://wiki.kerbalspaceprogram.com/w/images/7/73/KerbinDeltaVMap.png Use this map in conjunction with a performance calculator, like Engineer or MechJeb, to design your ships to your mission requirements.

While I may be taking the thread too seriously, most actual astrodynamics classes will do one of two things: A) Try to try teach it too early to a group of students who don't have the appropriate requisites yet, in which case you either fail everyone, or you teach a set of seemingly-unrelated equations without derivations and go heavy on the qualitative stuff like historical context and operations. I've seen this approach taken at shockingly-"good" institutions, by professors who should know better but treat it as no big deal. Really, you might learn more effectively from playing KSP. First, make sure your students have a basic grasp of vector calculus, conic and spherical geometry, frame transformations, and differential equations. Given that knowledge, astrodynamics itself is surprisingly trivial to teach from basic principles. I am a particularly big fan of Curtis's "Orbital Mechanics"; it has a very logical and scaffolded progression, it is well-written and easy to walk through (even by yourself), and it covers a pretty diverse array of topics in addition to the mechanics themselves (including numerical solutions, body kinematics, and launch trajectory modeling). Here's the obligatory Amazon link to the latest edition, in case you're curious: http://www.amazon.com/gp/aw/d/0080977472?pc_redir=1409138424&robot_redir=1

We've tried that a couple of times; it usually ends up on some form of command capsule plus oversized rocket engine--no fuel, no seperators, etc. This will usually be capped off by a full spread of eight launch supporters. Sometimes we'll get lucky and she'll remember a booster or two, but even that is never balanced and so the results are spectacular. I can't bring myself to "correct" her, as it's so much fun.

The potential of KSP as an educational family game is highly under-rated. My four-year-old daughter and I play together frequently, and she's fallen in love with the Kerbals. It's a great opportunity to explore both basic physics principles (gravity, drag, etc.) and even more advanced concepts (orbits, Karman line, apoapsis, transfers, etc.) in a very accessible and qualitative manner.

Last Exile!

That moment when you use all your transit stage fuel lining up the world's most beautiful aerobraking trajectory, only to realize that a) the frame transition now puts you at 20000 m periapsis altitude instead of 80000 m, you're inbound for Eve and are therefore unavoidably screwed, and c) your four-year-old daughter, who is in love with Jebediah, is about to watch him perform an extreme lithobraking maneuver. Better revert, quick!

Between the extensive sandbox nature of the game, it's educational nature, and whimsical character and art style, I think KSP is going to be around for years. It's already in a place where Sqiad's role is an implementer of official content; if (say) they dropped off the face of the earth, the player and mod communities would still be going just as strong. We might have to relocate the forum, but that's about it.

Compare the above video to the Musk tweet...

Period = 2 pi sqrt(a^3 / mu) So... ((Period / (2 pi))^2) mu)^(1/3) = a Basically, you need to divide by the gravitational parameter in the original expression... Keep in mind it has units of length-cubed-over-time-squared.

Numerical stability is also an issue. With restricted two-body, there's really only one numerical problem to solve at each time-step (anomaly), and there are very efficient algorithms for doing so with a controllable amount of error. When solving an n-body by straight integration (even a restricted two-body, actually, without the right conservative integrators), small errors (round-off in particular) will be exponentially magnified over time. Computing the EoMs are actually tricky in and of themselves, even ignoring the integration step, because of a) the way your problem scales (though this can be partly addressed by bounding knowledge of your problem size and by using dipoles for influence grouping), that nasty square root operator with which you are evaluating the magnitude of your relative range vectors, and c) the loss of precision you encounter when you sum large numbers of near-uniformly-distributed vectors, many of which will have at least one component opposed to it's brethren.

Absolutely, especially while you're young. It's all about investing on experience right now in you career. Try not to be unpaid, especially after you graduate, but if you can just get your foot in the industry door, do it. Even if the place folds after a year, their engineers don't just disappear--now you have a network of former co-workers with jobs in multiple companies and locations. I was working that particular internship for a very large corporation (one of the largest, on fact), and while there are interesting programs with similar companies that would be very fun to work on, I wouldn't go back to that particular corporation without a 10x pay raise. I would strongly recommend the following book. It's short, but has near-infinite value for professional engineers (aerospace and others). http://www.amazon.com/gp/aw/d/156347655X?cache=4d47d968c0050f1c48383eb79a2311f7π=SY200_QL40&qid=1408559511&sr=8-1#ref=mp_s_a_1_1

If autonomy is your goal, than consulting is your road, but you won't be able to take it without a PhD, MBA, and/or literally decades of experience as a SME (subject matter expert). Otherwise, the options you mentioned have equal degrees of autonomy in our field. Aerospace may, possibly, have more, because there is a chance you'll end up in a systems engineering and/or integration role, where you get to play a greater part in the design process instead of just building widgets to spec. My advice: don't focus on pure autonomy so much as visibility, in two senses of the meaning: First, visibility others have of your work and associated successes, which will give you more leverage in the long run to control what you work on. Second, visibility in that you can see and know what exactly you are contributing to and thereby remain inspiried and interested in the mission. I spent my first internship in the aerospace writing XML parsers for a data acquisition system three degrees removed from the stuff that would actually go into space. It was incredibly depressing--I got out of there as soon as I could.

Getting to Mars... Well, depending on who you talk to, we're either 7 or 70 years away from sending people. Let's break it down. The first difference is public vs. private. SpaceX (I.e., Musk) says he'll do it by 2025 or so, come hell or high water, even if government funding is as scarce as a donkey's shaved ass in an Amish strip club. You be the judge. For the serious among you, there's a lot of work to be done. The really sad part is, in my opinion, how little of it has really been done since the termination of the Apollo program. Nominally, ISS should be nothing but a big getting-technology-ready--for-Mars facility. In reality, it's turned into a big (though still useful, if only marginally so) session of Kumbaya-in-space. There are seriously big and real problems we still need to solve, like life-level radiation shielding and multi-year multi-person isolated self-sustaining space environments, that we could have solved if we had seriously started tackling them 30 years ago. I put this research, of fully-funded and properly managed, at 8-10 years. Since we are being realistic, though, that is more likely multiplied by a factor of 2 or 3. Then there's the issue of space hardware--there and back again, if you will. This problem is the usual scapegoat, but a minor investment in orbital assembly technologies would have fixed this in as little as 3 years using existing launch vehicles. A sustained evolved shuttle pyrenean would have completely circumvented the issue of manned rating, and you could have skipped the process and used existing HLVs for every other component at a near-zero NRE, and you could start right away. Descent/Ascent is another problem entirely. Fellow KSP players, you are more familiar with the details of the problem than most. We can't even use the Magic Bouncy Ball approach past 800 kg or so, and Sky Crane approach will never be approved for manned missions (far too risky, far too many likely and catastrophic failure modes). If you really want to enable there-and-back Mars missions, this is where your efforts will be focused, and why the recent LDSD test was one of the first indicators that NASA is actually getting serious about at least half of this problem. Ascent and return, at least, can be solved by automated and pre-placed transit of fuel repositories. I'd put these technologies at 8-10 years. The biggest problem that remains, then, is the question of what to do while we're spending months and months waiting for a fuel-efficient return. Should we drill? Rove via buggy? Focus on the poles? Strangely, few people seem to have seriously thought about this issue, aside from the seed appeal of putting people there is the first place. It will probably be left up to the geologists, like Apollo, and the astrobiologists, which isn't entirely a bad thing. Still, I think some inputs from us engineers wouldn't be inappropriate. Another round of golf-off-Earth, anyone?

A word of caution... While 130 m/s is a good threshold, you need to reach that state at a sufficient rate of acceleration, or you will have an unstable pitchover. For example, try to reach it within the first (say) 20 seconds of your flight (though that number is arbitrary, I can't crunch the numbers right now). You could also use T/W as a criteria, but this will require the Engineering add-on.

RIC vectors in the nav ball? I'm so excited! I can finally see my orbital plane n spherical space. Now can we add plane contours for targets, as well?

Curious question arose in my red-wine-sauteed brain as I renewed my membership... What is the crossover between KSP and the Planetary Society? Any other members here? Surely I'm not the only one. Random, I know... I blame the 2011 Toscana and a desire to become more familiar with my beloved KSP community.

Landing multiple gravity assists (not by planning but purely by taking advantage of noticed opportunities) in a row...

My advice: - Don't bother seeing more than one or two IMAX films; most of them are just long drawn-out sequences of liftoff, over and over and over... Worth seeing, but after a couple of iterations, three will be better ways to spend your time. - Allocate a couple hours for just wandering around the rocket garden with your mouth gaping open on awe. About half that time will be spent starting at an F-1. - Go on every tour you can. My particular favorite is the "Then and Now" tour, which takes you (along other things) to some of the very first original pads and bunkers. Very, very cool. - Follow the advice from above--schedule your visit around a launch, if you can. The simulators are also neat. - You have to see the Atlantis. I just saw the Endeavour up close for the first time today, and it was more amazing than I ever could have hoped. There I was, a grown man, standing under the engines and looking up with literal tears in my eyes.

Yeah, baby, yeah! KSP is definitely in my top five--probably fourth, to be specific--favorite games of all time.